Throughout this report we will investigate many parameters on income. We are utilizing the American Community Survey (ACS), which was conducted to represent approximately 3.5 million households per year. Here we will investigate the specific attributes of Location, Race,Sex,Work Hours,Travel Time to Work and their affects on Person’s Total Income. We will show patters observed along with several conclusions from the data.

image

Average Total Person Income according to States of Residence

Average Total Person Income according to Birthplace

People live in District of Columbia have the highest average total person income. It is then followed by people from Connecticut, Maryland, New Jersey and Massachusetts. The similar pattern can be found in the plot of average total person income according to birthplace except that Maryland is replace by New York and it moves one place up the list. District of Columbia still tops the list. In general, people born in the following five places, District of Columbia, Connecticut, Maryland, New York, Massachusetts, are also those who have highest average total person income in adulthoods.

How does gender influence income?

From the plot, we can see that:
- 1. both density distributions of males’ income and females’ income are left-skewed and have a long tail.
- 2. There are more fluctuations in females’ income.
- 3. Generally speaking, males have higher income than females.

How does age influence income?

Then we draw a boxplot to show the income of every age.

We see things are similar with the previous plot.

Density Plots of Minutes of Travel to Work, Work Hours Per Week and Personal Income

Here we will look at many density plots for comparsions between races and the total population. On the race plots you will see “Pop” markders, which idenfiy the Median for the overall population. Where we see “Race” signifies the median of that particular race.


Overall the plots help show the differences between the Population Median and each Race Median. This can give us an idea of what we can expect to see in the linear regression in terms of correlations between Race,Sex,Work Hours,Travel Time to Work and a Person’s Total Income.

While the Work Hours Per Week are very consistent, there are differences between Races in Travel Time to Work. Native Americans and Alaskans were well below hte Population Median. This could be a potential relatioship between living on reservations or in isolated areas, where work is mostly local. The Asian population however traveled farther to work that the US Population Median.

In terms of Total Personal Income all of the races were less than the US Population Median, except White and Asian.

We can also see from the plots the differences in Sex between racial populations. One interesting observation is that there is a certain point in almost all races of the Personal Income, where to the left of the point, Female density is higher. Everything to the right of that same point, the Male density is higher.

Linear Regression Analysis of Race,Sex,Work Hours,Travel Time to Work on a Person’s Total Income

summary(svyglm(PINCP ~ SEX + JWMNP + WKHP + RAC1P, design = acsdesign))
Estimate Std. Error tvalue Pr(>|t|)
Intercept -4.89E+03 2.65E+01 -184.79 <2e-16 ***
Female -1.12E+04 1.16E+01 -961.33 <2e-16 ***
Travel Time 1.58E+02 3.08E-01 512.23 <2e-16 ***
Work Hours 1.47E+03 5.98E-01 2465.68 <2e-16 ***
Black -1.30E+04 1.56E+01 -836.15 <2e-16 ***
American Indian -1.55E+04 4.32E+01 -359.6 <2e-16 ***
Alaska Native -1.04E+04 3.50E+02 -29.56 <2e-16 ***
American and Alaskan -1.80E+04 1.09E+02 -165.79 <2e-16 ***
Asian 7.33E+03 2.94E+01 249.22 <2e-16 ***
Native Hawaiian -1.33E+04 1.36E+02 -97.53 <2e-16 ***
Some Other Alone -2.21E+04 2.14E+01 -1031.67 <2e-16 ***
Two or More -9.35E+03 3.46E+01 -270.36 <2e-16 ***
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘’ 1


As we can see from the results of the linear regression there are many significant linear relationships between the Race,Sex,Work Hours,Travel Time to Work and a Person’s Total Income.

Even within the news today that are large discussions about the issues of inequality in pay for gender.The linear regression showed that there is a negative correlation between Females and Person’s Total Income, which we saw from the previous analysis.

These tables also point out the differences we see for Race and Person’s Total Income. Comparing to the White popultation of the United States there is a negative linear correlation for all races, except Asian. We saw this within the plots above where the Asian population had a higher Median thatn the overall population in Total Personal Income and interestingly was the only race higher a median higher than the population median for Travel Time to Work.